\begin{table}[H] \centering \vspace*{-2.5cm}   \caption{Treatment effects on recent use of online resources and contact with an organization when responding to domestic and sexual violence before COVID-19}   \label{tab:rf_13} \scriptsize \begin{tabular}{@{\extracolsep{0pt}}lccc} \hline \\[-1.8ex] \\[-0.5ex] \multicolumn{4}{l}{\textbf{Panel A: Controlling by the lagged dependent variable and covariates selected by LASSO}} \\ \hline \\[-1ex]  & \shortstack{Index of \\ (1,1) } & \shortstack{Used online \\ resources} & \shortstack{Contacted \\ organization} \\ \\[-1.8ex] & (1) & (2) & (3)\\ \hline \\[-1.8ex]  SM Individual & 0.017 & 0.036 & $-$0.006 \\   & ($-$0.044, 0.078) & ($-$0.017, 0.090) & ($-$0.049, 0.038) \\   & p = 0.586 & p = 0.185 & p = 0.802 \\   & & & \\  SM Group & 0.032 & 0.017 & 0.023 \\   & ($-$0.034, 0.098) & ($-$0.041, 0.076) & ($-$0.024, 0.070) \\   & p = 0.346 & p = 0.561 & p = 0.343 \\   & & & \\  TV & 0.028 & 0.025 & 0.013 \\   & ($-$0.038, 0.094) & ($-$0.034, 0.083) & ($-$0.034, 0.060) \\   & p = 0.409 & p = 0.405 & p = 0.598 \\   & & & \\ \hline \\[-1.8ex] SM Individual = SM Group \\(p-value) & 0.6573 & 0.5232 & 0.2372 \\ SM Individual = TV \\(p-value) & 0.7471 & 0.6963 & 0.449 \\ SM Group= TV \\(p-value) & 0.906 & 0.8077 & 0.6784 \\ Num. Lasso covariates & 8 & 11 & 7 \\ R$^{2}$ & 0.468 & 0.498 & 0.295 \\ \hline \\[-0.5ex] \multicolumn{4}{l}{\textbf{Panel B: Controlling by the dependent variable at baseline (if available) }} \\ \hline \\[-1ex] SM Individual & 0.010 & 0.035 & $-$0.012 \\   & ($-$0.051, 0.071) & ($-$0.019, 0.089) & ($-$0.056, 0.031) \\   & p = 0.747 & p = 0.211 & p = 0.578 \\   & & & \\  SM Group & 0.025 & 0.016 & 0.020 \\   & ($-$0.041, 0.092) & ($-$0.043, 0.075) & ($-$0.027, 0.068) \\   & p = 0.456 & p = 0.604 & p = 0.399 \\   & & & \\  TV & 0.024 & 0.027 & 0.011 \\   & ($-$0.042, 0.090) & ($-$0.031, 0.086) & ($-$0.036, 0.059) \\   & p = 0.473 & p = 0.361 & p = 0.635 \\   & & & \\ \hline \\[-1.8ex] SM Individual = SM Group \\(p-value) & 0.6531 & 0.528 & 0.175 \\ SM Individual = TV \\(p-value) & 0.676 & 0.8101 & 0.3251 \\ SM Group= TV \\(p-value) & 0.9755 & 0.7017 & 0.7165 \\ R$^{2}$ & 0.459 & 0.489 & 0.280 \\ \hline \\[-0.5ex] \multicolumn{4}{l}{\textbf{Panel C: No covariates }} \\ \hline \\[-1ex] SM Individual & 0.005 & 0.031 & $-$0.014 \\   & ($-$0.058, 0.068) & ($-$0.024, 0.086) & ($-$0.058, 0.030) \\   & p = 0.887 & p = 0.265 & p = 0.538 \\   & & & \\  SM Group & 0.036 & 0.022 & 0.025 \\   & ($-$0.033, 0.104) & ($-$0.038, 0.082) & ($-$0.023, 0.073) \\   & p = 0.308 & p = 0.480 & p = 0.312 \\   & & & \\  TV & 0.043 & 0.036 & 0.021 \\   & ($-$0.025, 0.111) & ($-$0.024, 0.095) & ($-$0.027, 0.069) \\   & p = 0.214 & p = 0.241 & p = 0.394 \\   & & & \\ \hline \\[-1.8ex] Control Mean & -0.09 & 1.342 & 1.138 \\ SM Individual = SM Group \\ (p-value) & 0.3732 & 0.7507 & 0.1142 \\ SM Individual = TV \\(p-value) & 0.2684 & 0.8848 & 0.1567 \\ SM Group= TV \\(p-value) & 0.8326 & 0.6511 & 0.8733 \\ Observations & 4,165 & 4,165 & 4,165 \\ R$^{2}$ & 0.424 & 0.471 & 0.255 \\ \hline \hline \\[-1.8ex] \multicolumn{4}{l} {\parbox[t]{10.5cm}{ \textit{Notes:}
  We report estimates from WGLS regressions where the weights are in the inverse probability of treatment 
  assignment, including randomization block fixed effects. 
  Regressions in Panel A use as controls the covariates selected by LASSO in which the treatment indicators,
  lagged dependent variable, and fixed effects are forced into model and covariates are selected from the outcome family.
  Regressions in Panel B include the dependent variable at baseline (if available) as a control. 
  Regressions in Panel C do not include any variable as a control. 
  95\4 confidence intervals are in parenthesis (due to two-sided t-tests). 
  * denotes p$<$0.1, ** denotes p$<$0.05, and *** denotes p$<$0.01.}} \\\end{tabular} \end{table} 